< >

Carola said

I have read that while two-stage least squares and other instrumental

variables estimators are consistent, they are not unbiased in small

samples. I wonder how large should the sample be in order to safely

use two-stages least squares/instrumental variable estimation. Is

there a cutoff N?

As any textbook treatment of IV will demonstrate, IV estimators are

biased at any sample size that you are likely to encounter in

empirical work. The amount of the bias may become negligible, but it

is there. You could evaluate the degree of bias with Monte Carlo

readily enough.

Kit Baum, Boston College Economics and DIW Berlin

http://ideas.repec.org/e/pba1.htmlAn Introduction to Modern Econometrics Using Stata:

http://www.stata-press.com/books/imeus.html*

* For searches and help try:

*

http://www.stata.com/help.cgi?search*

http://www.stata.com/support/statalist/faq*

http://www.ats.ucla.edu/stat/stata/